Presentation to MSU Department of Psychology, Program Evaluation Brown Bag Series, East Lansing, MI
Center for Statistical Training and Consulting
2024-12-05
Missing data (MD) is a common issue afflicting research and evaluation studies.
Warning
If you do much evaluation work, you will run into missing data.
Any measurement you intended to obtain, but did not get.
Handling missing data well enacts our guiding principles[1]:
MCAR is when neither observed nor unobserved variables predict which data is missing
What are the consequences of missing data?
In appropriate handling of missing data can cause analyses to yield biased results.
Misalignment of analyzed sample and intended population
Most statistical software defaults to listwise deletion of cases that have any missing values on the variables involved in an analysis. That reduces statistical
An ounce of prevention is better than a pound of cure
[2]